We have continued our studies on the role of tubulin in apoptosis. All of the antimitotic drugs investigated so far have induced apoptosis in a variety of cell types. While there are a number of possible loci of action for these drugs, they can release cytochrome c from isolated mitochondria (a sign of apoptosis) which contain tubulin presumably in the outer membrane. This finding simplifies looking for a mechanism. The samples of our mitochondrial complex containing tubulin and associated proteins were lost by a proteomics company which subsequently went bankrupt. Thus we have analyzed possible direct interactions of certain pro- and anti- apoptotic proteins with tubulin. A direct interaction between tubulin and several pro-apoptotic and anti-apoptotic members of the Bcl-2 family has been demonstrated by effects on the assembly of microtubules from pure rat brain tubulin. Bcl-2, Bid and Bad inhibit assembly sub-stoichiometrically, whereas peptides from Bak and Bax promote tubulin polymerization at near stoichiometric concentrations. These opposite effects on microtubule assembly are mutually antagonistic and are not mediated by the BH3 homology domains common to all members of the family. Pelleting experiments with paclitaxel-stabilized microtubules show that Bax is associated with the microtubule pellet, whereas Bid remains primarily with the unpolymerized fraction. Moreover, these interactions require the presence of the highly acidic C-termini of alpha- and betas-tubulin as they do not occur with tubulin S in which the C-termini have been removed proteolytically. We are now in the process of determining whether the antimitotic drugs of the colchicine family, vinca alkaloids, and paclitaxel can interrupt the association between tubulin and members of the Bcl-2 family using plasmon resonance techniques in collaboration with Peter Schuck and Inna Gorshikova (ORS). Finally it is planned to look again for these proteins, as well as VDAC, hexokinase, etc in the high molecular weight mitochondrial complex. We are also continuing the analysis of the thiol groups of tubulin in collaboration with Yeounjin Kin (Mass Spec Facility, NIDDK) and Dan Sackett (NICHD). The reaction of tubulin thiols with N-ethyl maleimide (NEM) under first-order conditions is studied as a function of time, and tryptic peptides were analyzed by HPLC/MS-MS methods. Reaction rates of 10 of the 20 thiols could be measured ? the remaining 10 defied analysis, because peptides contained several thiols, etc. The reaction of alpha-tubulin Cys 347 was too fast to determine a rate and was assigned a rate constant of >0.23 s-1. betaC-241 and -356 were very fast (k = 1.8 and 2.5 10-2 s-1 resp.) , four thiols were an order of magnitute slower and three others were very slow to react. We are currently exploring the local environmental basis for these rates and are trying to determine which of the thiols has the most profound effect on polymerization competence.